The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2008, 181, 5451 -5461
Copyright © 2008 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow A correction has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fallang, L.-E.
Right arrow Articles by Sollid, L. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fallang, L.-E.
Right arrow Articles by Sollid, L. M.
Right arrowPubmed/NCBI databases
Medline Plus Health Information
*Autoimmune Diseases

Complexes of Two Cohorts of CLIP Peptides and HLA-DQ2 of the Autoimmune DR3-DQ2 Haplotype Are Poor Substrates for HLA-DM1

Lars-Egil Fallang*, Sujin Roh{ddagger}, Anders Holm*, Elin Bergseng{dagger}, Taejin Yoon{ddagger}, Burkhard Fleckenstein*, Arunima Bandyopadhyay{ddagger}, Elizabeth D. Mellins2,3,{ddagger} and Ludvig M. Sollid2,3,*,{dagger}

* Centre for Immune Regulation and Institute of Immunology, University of Oslo, Oslo, Norway; {dagger} Centre for Immune Regulation and Institute of Immunology, University Hospital, Oslo, Norway; and {ddagger} Department of Pediatrics, Stanford University, Stanford, CA 94305


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Atypical invariant chain (Ii) CLIP fragments (CLIP2) have been found in association with HLA-DQ2 (DQ2) purified from cell lysates. We mapped the binding register of CLIP2 (Ii 96–104) to DQ2 and found proline at the P1 position, in contrast to the canonical CLIP1 (Ii 83–101) register with methionine at P1. CLIP1/2 peptides are the predominant peptide species, even for DQ2 from HLA-DM (DM)-expressing cells. We hypothesized that DQ2-CLIP1/2 might be poor substrates for DM. We measured DM-mediated exchange of CLIP and other peptides for high-affinity indicator peptides and found it is inefficient for DQ2. DM-DQ-binding and DM chaperone effects on conformation and levels of DQ are also reduced for DQ2, compared with DQ1. We suggest that the unusual interaction of DQ2 with Ii and DM may provide a basis for the known disease associations of DQ2.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Major histocompatibility complex class II molecules are {alpha}β-dimeric membrane glycoproteins expressed on the surface of APCs of the immune system. Their function is to present peptide Ags derived from endosomal proteins to CD4+ T cells. To ensure that newly synthesized class II molecules intersect with and bind endocytosed peptides, the {alpha}β dimer associates with the chaperone molecule invariant chain (Ii),4 which both prevents premature binding of ligands and promotes the entrance of the class II-Ii complexes into the endocytic pathway (1).

Ii is progressively degraded by proteases in the endocytic pathway, leaving CLIP bound to the binding groove of all previously studied MHC class II molecules. The subsequent release of CLIP in late endosomes is required for antigenic peptide binding, a process known to be accelerated by HLA-DM (DM). The MHC II-peptide complex is then transported to the cell surface for inspection by CD4+ T cells (2).

Previous data on peptides eluted from HLA-DQ2 (DQ2) molecules (DQA1*0501, DQB1*0201) that were affinity purified from B lymphoblastoid cell lines (B-LCL) revealed the presence of large amounts of Ii-derived peptides (54% of total) (3, 4). Intriguingly, the population of Ii peptides was dominated by a unique peptide species, here termed CLIP2, which had lower IC50 values/higher affinity for DQ2 when compared with the conventional CLIP (CLIP1) peptides (3, 4). DQ2 is part of the "autoimmune" HLA DR3-DQ2 haplotype associated with multiple autoimmune diseases (5, 6). Motivated by this fact, we investigated the binding of CLIP2 to DQ2 and asked whether the presence of substantial amounts of CLIP peptides reflected altered interaction of DQ2 with DM, in comparison to DM interaction with HLA-DR3 (DR3) and HLA-DQ1 (DQ1).


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Cell lines

The following EBV-transformed B lymphoblastoid cell lines (B-LCL) were used in this study: CD114 (from a celiac disease patient) expressing DQ2 and DR3; 2.2.93 expressing DP4, DQ1, DR1, DR3, and no DM (7); 8.1.6 expressing DP4, DQ2, and DR3 (8); 9.5.3 expressing DP4, DQ2, DR3, and no DM (8); 3.1.3 expressing DP4 and DQ1 (9); and 9.22.3 expressing DP4 and DQ2 (8). The DR3-expressing cells also express DR52a. The DM transfectants of 2.2.93 and 9.5.3 (2.2.93-DM and 9.5.3-DM, respectively) were generated by retroviral transduction as previously described (7).

Antibodies

Abs used in this study were B8.11 (IgG2b, anti-DR, gift from B. Malissen (Centre d’immunologie de Marseille Luminy, Marseille, France)), L243 (IgG2a, anti-DR), ISCR3 (IgG2b, anti-DR), SPV-L3 (IgG2a, anti-DQ, gift from H. Spits (University of Amsterdam, Amsterdam, The Netherlands)), Ia3 (IgG2a, anti-DQ; Biodesign), 2.12.E11 (IgG1, anti-DQ2), B7/21.2 (IgG3, anti-DP), XD5.A11 (IgG1, anti-class II, gift from P. Cresswell (Yale University School of Medicine, New Haven, CT)), 5C1 (IgG1, anti-DM), 16.23 (IgG3, anti-DR3), anti-HLA-DM conjugated with PE (IgG1; BD Pharmingen), CerCLIP (IgG1, anti-human CLIP, gift from P. Cresswell).

Flow cytometry

For cell surface staining, cells were incubated (40 min on ice) with primary Abs and washed. Bound Ab was detected by incubation (40 min on ice) with goat F(ab')2 anti-mouse IgG (H + L) conjugated with FITC or PE (Caltag Laboratories). For intracellular staining, cells were fixed and permeabilized using the Cytofix/Cytoperm kit (BD Pharmingen) and then stained using PE-conjugated Ab. Cells were analyzed using a FACScan flow cytometer (BD Biosciences) and data were analyzed using FlowJo software (Tree Star).

Pulse-chase immunoprecipitation

Cells were washed three times and starved for 1–2 h in methionine/cysteine-free RPMI 1640 containing 10% dialyzed FBS (Invitrogen). Cells were pulsed with 0.1 µCi/ml ExpreSS 35S labeling mix (PerkinElmer) for the indicated times, then washed and chased in complete RPMI 1640 containing 10% FBS and 2 mM L-glutamine (37°C, 5% CO2). Aliquots of cells were collected and washed at the indicated time points and lysed in buffer (Tris-HCl (pH 8.0) with MgCl2, 1% Nonidet P-40, and complete protease inhibitors (Roche Diagnostics)) at 4°C for 1–2 h. Lysates were precleared three times with normal mouse serum and Pansorbin (Calbiochem) and once with protein A-Sepharose beads (Amersham Biosciences), then normalized based on total radioactivity measured by a beta counter (Wallac). Immunoprecipitations were performed by incubating the normalized lysates with protein A-Sepharose beads conjugated with class II-specific Abs overnight at 4°C. Proteins were eluted by boiling the precipitates in reducing SDS sample buffer containing 62.5 mM Tris-HCl (pH 6.8), 1% SDS, 3% glycerol, 0.007% bromphenol blue, and 1% 2-ME and then separated by SDS-PAGE. Bands were visualized by exposing dried gels to radiography films (Kodak).

Generation of soluble (s) DQ2- and DR3-peptide complexes

Water sDQ2 engineered to have a peptide ligand tethered to the β-chain via a thrombin cleavable linker and a Fos-Jun leucine zipper pair replacing the transmembrane domains, was expressed in insect cells using a baculovirus expression vector system (10). The sDQ2 molecules with deamidated {alpha}I gliadin peptide (QLQPFPQPELPY) and deamidated {gamma}I gliadin peptide (PEQPQQSFPEQERP) have been previously described (10). The insertion of CLIP1 and CLIP2 sequences (PVSKMRMATPLLMQA and MATPLLMQALPMGAL, respectively), into the DQ2 β-chain construct were completed by ligation of hybridized oligonucleotides (corresponding to the peptide sequences) with single-strand overhangs complementary to those of the MroI/NarI restriction sites of the original β-chain construct. The sDQ2-CLIP1 M91A and sDQ2-CLIP M91P complexes were constructed by performing site-directed mutagenesis of the sDQ2-CLIP1 plasmid using primers containing the desired nucleotide change (QuikChange Multi Site-Directed Mutagenesis Kit; Stratagene) according to the manufacturer’s instructions. The water sDR3-CLIP1 construct was created in a similar fashion as the original sDQ2 construct. cDNA was synthesized by RT-PCR (Superscript II; Invitrogen) from mRNA (RNeasy; Qiagen) isolated from the 8.1.6 cell line according to the manufacturers’ instructions. The DRA1*0101 and DRB1*0301 sequences were PCR amplified from cDNA, where the DRA1*0101 primer contained an N-terminal BglII restriction site. The Jun-Fos leucine zippers from the sDQ2-CLIP1 construct containing C-terminal BglII (DQ{alpha} chain) or BamHI (DQβ chain) restriction sites were PCR amplified with primers containing DR{alpha} or DRβ complimentary tails, allowing for megaprimer PCR attachment to the C-terminal ends of the DR{alpha} (Fos) and DRβ (Jun) constructs. The N-terminal end of the sDQ2-CLIP1 molecule containing the BamHI restriction site, leader sequence, CLIP1, and linkers was PCR amplified with primers containing DRβ complimentary tails, allowing for megaprimer PCR attachment to the N-terminal end of the DRβ chain. The constructs were cloned into the pAcAB3 vector by means of BamHI (DRβ) and BglII (DR{alpha}) restriction enzymes. The oligonucleotides used for the construction of the molecules can be found in Table I. Constructs were verified by DNA sequencing. The sDQ2 and sDR3 molecules were affinity purified using the mAbs 2.12.E11 or B8.11, respectively.


View this table:
[in this window]
[in a new window]

  		Table I. Oligonucleotides used in the construction of sDQ2-peptide and sDR3-CLIP1 complexes

 
Generation of sDM

sDM molecules were produced in stably transfected S2 cells and purified by immunoaffinity chromatography and size exclusion chromatography (11).

Fluid-phase peptide-binding assay

Detergent-solubilized DQ2 and DR3 molecules were purified from the CD114 cell line and used in a fluid-phase competitive inhibition assay with 125I-labeled indicator peptides as previously described (12). Briefly, labeled indicator peptides (KPLLIIAEDVEGEY; MB 65-kDa Hsp 243–255Y or EPRAPWIEQEGPEYW; HLA class I {alpha} 46–60) for DQ2 and KTIAYDEEARR; MT 65-kDa 3–13 for DR3) and unlabeled peptides were incubated with DQ2 or DR3 molecules overnight at 37°C in a pH 4.9 citrate phosphate buffer. Complexes of peptide and DQ2/DR3 molecules were separated from unbound peptides by spin column chromatography. Radioactivity was determined, and the concentrations of the competing peptides required for half-maximal inhibition of the binding of the indicator peptide (IC50) were calculated.

Peptide-binding assays measuring the DM effect using water-soluble peptide-linked molecules were done in a similar manner. Molecules (2 µg) were treated with Formula U of thrombin (Novagen) for 90 min at room temperature followed by inhibition with 1 mM 4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride (Sigma-Aldrich). The cleaved molecules were incubated overnight at 37°C in a pH 4.3 citrate phosphate buffer in the presence or absence of various amounts of sDM (as indicated in the figures) and, in some experiments, noncleaved molecules (0–7 µM) as well, along with the 125I-labeled indicator peptides and protease inhibitors.

Solid-phase peptide exchange assay

HLA class II molecules captured from cell lysates were used for a peptide exchange assay (13). Briefly, clarified cell lysates of 0.5 x 106 cells of B-LCL 8.1.6 and 9.5.3 were added to wells coated with either mAb 2.12.E11 (anti-DQ2), or mAb L243 (anti-DR). After incubation at 4°C overnight, the wells were washed, and 2.5 µM biotinylated indicator peptides were added (identical peptide sequences as in fluid-phase binding assay) in a pH 4.3 citrate phosphate buffer containing protease inhibitors. To adjust for the amount of captured DQ2 and DR3, biotinylated mAb B8.11 (anti-DR) or SPV-L3 (anti-DQ) in pH 7.4 citrate phosphate buffer was added to the relevant wells and incubated for 48 h at 37°C. The wells were washed and streptavidin-europium diluted 1/2000 in assay buffer (Wallac) was added, incubating for 60 min at 37°C. The wells were washed and enhancement solution (Wallac) was added, incubating for 20 min before measurement in a time-resolved fluorometer (1234; Wallac).

Dissociation experiments of peptides from water sDQ2 and sDR3

Water-soluble DQ2-CLIP1, DQ2-CLIP2, DQ2-CLIP1 M91A, DQ2-CLIP1 M91P, DQ2-{alpha}I gliadin, DQ2-{gamma}I gliadin, and DR3-CLIP1 molecules (1.6 µM) were treated with thrombin (1 U per 32 µg) for 90 min at room temperature, followed by 4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride treatment (1 mM). The cleaved molecules were incubated at 37°C in the presence or absence of 6 µM sDM in a pH 5.0 citrate phosphate buffer for the required length of time. High-affinity competitor peptides were added in excess (35 µM) (MB 65-kDa Hsp 243–255Y for DQ2 and MT 65-kDa 3–13 for DR3). The dissociation was stopped by neutralizing the sample with cold citrate phosphate buffer (pH 7.2). Released peptides (CLIP1: RDSGPVSKMRMATPLLMQAGAGSLVPR, CLIP2: RDSGMATPLLMQALPM GALGAGSLVPR, CLIP1 M91A: RDSGPVSKARMATPLLMQAGAGSLVPR, CLIP1 M91P: RDSGPVSKPRMATPLLMQAGAGSLVPR {alpha}I gliadin: RDSGQLQPFPQPELPYGAGSLVPR, and {gamma}I gliadin: RDSGPEQPQQSFPEQERPGAGSLVPR) were removed by filtration using 10,000 MWCO microcon centrifugal filters (Millipore), and the bound peptides were eluted with 0.1% TFA/2 mM DTT in the presence of indicator peptides (16.1 pmol of each) for 30 min at 37°C. The indicator peptides have identical sequence to the released peptides except for a serine-threonine substitution in position 3. Peptides were isolated using microcon filters, lyophilized, and treated with 0.1 M trimethylsilyl bromide in 100% acetonitrile (ACN) for 60 min at room temperature. Following lyophilization, the peptides were reconstituted in 0.1% TFA and purified over Poros 20 R2 reversed-phase particles (Applied Biosystems) packed in GELoader tips (Eppendorf) and eluted directly onto a stainless steel target plate (Bruker Daltonics) using ACN/0.1% TFA (70/30, v/v) containing 5 mg/ml {alpha}-cyano-4-hydroxycinnamic acid. After crystallization, the samples were analyzed on a MALDI-TOF/TOF mass spectrometer (Ultraflex II; Bruker Daltonics). The relative quantification of the eluted peptides was performed by comparing the first two peaks of the isotopic distribution of the eluted peptides with the corresponding peaks of the indicator peptide, which has a +14-Da mass increase due to the serine-threonine substitution. The dissociation kinetics were fit into the single exponential decay function (Y = As x exp(–ksx)) using GraphPad Prism (version 3.02).

Isolation of peptides bound to DQ2 and DR3 molecules from 8.1.6 and 9.5.3 cells

DQ2 and DR3 molecules were affinity purified from the 8.1.6 and 9.5.3 B-LCLs using anti-DQ2 mAb 2.12.E11 and anti-DR mAb B8.11 essentially as previously described (12). Purified molecules (50 µg) were washed with Milli-Q water using 10.000 MWCO microcon filters, followed by acid elution (0.1% TFA) of the peptides. The peptides were isolated using microcon filters and lyophilized. Sequencing of the peptides was performed by reconstituting the peptides in 10 mM formic acid and separating on a series 1100 nano-liquid chromatography (LC) instrument (Agilent) coupled to an ion trap mass spectrometer (Esquire 3000+; Bruker Daltonics). Quantification of the eluted peptides was executed by LC-MS using an Isotope Coded Protein Labeling (ICPL) Kit (Serva) according to the manufacturer’s instructions. Briefly, lyophilized peptides and 75 pmol of each of six synthesized Ii peptides (Ii 81–104, Ii 81–103, Ii 82–103, Ii 81–101, Ii 93–109, Ii 93–108; all peptides synthesized by EZ-Biolabs) were dissolved in 4 µl of 50 mM NaHCO3 (pH 8.5) buffer and incubated for 120 min at 25°C with 0.6 µl of 12C-Nic-reagent or 13C-Nic-reagent, respectively. Subsequently, 0.7 µl of stop solution was added, and the samples were incubated for 20 min at 25°C before combining. The samples were treated with pH 11.9 phosphate buffer for 20 min at 25°C, neutralized with a citrate phosphate buffer to pH 7.2, and prepurified by HPLC (Column: Zorbax 300SB-C18 3.5Micron; Agilent), collecting the Ii peptides. The collected peptides were then separated on a series 1100 nano-LC instrument coupled to a Q-TOF mass spectrometer (microTOF-Q; Bruker Daltonics). Quantification of Ii peptides was completed by comparing the peak intensities observed in the mass spectrometer of the light and heavy ICPL-labeled peptides.

In vitro DM association assay

Cell pellets were lysed in buffer (50 mM Tris-HCl (pH 8.0) with 150 mM NaCl, 2 mM EDTA, 1% CHAPS, and complete protease inhibitors; Roche Diagnostics). Protein concentration was measured by Bradford assay, and class II molecules were immunoprecipitated from titrated amounts of lysate by overnight incubation with anti-DQ mAb (SPV-L3)-conjugated protein G-Sepharose beads. After washing with D-PBS containing 0.01% CHAPS, 1 ml of reaction buffer (1x PBS and 0.2% CHAPS) and 1–4 µg of sDM were added and then rotated for 3 h at 37°C. After washing with D-PBS containing 0.01% CHAPS, samples were boiled for 10 min with nonreducing sample buffer, cooled, and separated by SDS-PAGE and then transferred onto Immobilon polyvinylidene difluoride membrane (Millipore). The amount of DQ1/DQ2 and the associated sDM was detected using anti-class II β-chain mAb (XD5.A11) and anti-DM{alpha}-specific Ab (5C1), respectively. Ab binding was detected with HRP-conjugated goat anti-mouse followed by an ECL substrate (Amersham Biosciences).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
In cells, CLIP is released from DR3 more efficiently than from DQ2

Based on previous work showing the presence of significant amounts of CLIP peptides in elution profiles from DQ2-expressing cells (3, 4), we examined the relative stability of DR3-CLIP and DQ2-CLIP in cells by pulse-chase immunoprecipitation with Abs that bind to DR3 or DQ2 both in the presence and absence of CLIP peptides. To evaluate the influence of the level of DM, we used a panel of related cells expressing variable amounts of DM. Intracellular FACS staining revealed a hierarchy in DM levels, with 9.5.3 cells < 8.1.6 cells < 9.5.3-DM transfectant cells (Fig. 1A). In the absence of DM (9.5.3), CLIP was detected in association with both DQ2 and DR3 molecules at the 8-h postsynthesis time point. In the presence of DM (8.1.6 cells), the CLIP:DR3 ratio is substantially lower at 8 h of chase (23% of 9.5.3 ratio), whereas the change in the CLIP:DQ2 ratio is more modest (80% of 9.5.3 ratio; Fig. 1B and densitometric analysis in C). Thus, DQ2-CLIP complexes appear more resistant than DR3-CLIP to DM-mediated peptide exchange. This resistance can be overcome, as increased DM (9.5.3-DM) reduces DQ2-CLIP levels at 8 h (31% of 9.5.3 ratio). At all levels of DM, CLIP removal from DR3 was more efficient than from DQ2, however (Fig. 1, B and C).


Figure 1
View larger version (39K):
[in this window]
[in a new window]

  		FIGURE 1. The efficiency of Ii peptide release from DQ2 is lower than from DR3 and related to the level of HLA-DM. A, Intracellular DM staining of the indicated cell lines using anti-HLA-DM dimer Ab. B, 9.5.3, 8.1.6, and DM transfectant of 9.5.3 (9.5.3-DM) cells were pulsed for 1 h with [35S]methionine/cysteine and then chased in label-free medium. Cell lysates (3 x 106 cell equivalents/lane) were immunoprecipitated with anti-DQ mAb, SPV-L3 (top), or anti-DR mAb L243 (bottom) at the indicated times (hours) of chase and then analyzed by SDS-PAGE. Representative images from one of two independent experiments are shown. C, Densitometry analysis calculating ratios of CLIP to class II band intensity at 8 h from images shown in B. D, Assessment of cell surface CLIP-class II accumulation by FACS analysis. Indicated cell lines (9.5.3: DR3, DQ2, DP4, DM-null; 9.22.3: DQ2, DP4, DM+; 3.1.3: DQ1, DP4, DM+; 5.2.4: DP4, DM-null) were stained with anti-CLIP (CerCLIP.1; left) and anti-DQ (SPVL-3; right panel) followed by a secondary Ab (FITC-conjugated goat anti-mouse IgG). Control staining (without primary Ab: dotted lines) is comparable on all cell lines and histogram for one representative cell line is shown in each panel. CLIP:DQ ratios were calculated as MFICerCLIP– MFIisotype ÷ MFISPVL#– MFIisotype: CLIP:DQ1 = 1:108; CLIP:DQ2 = 1:108 p < 0.04. Representative data from one of more than three experiments are shown.

 
DQ2-CLIP complexes are surface expressed in DM-expressing cells

The finding of DQ2-CLIP complexes persisting during pulse-chase experiments leaves open the possibility that we are detecting mostly intracellular (predominantly pre-DM) complexes Therefore, we asked whether DQ2-CLIP complexes are also found at the cell surface in DM-expressing cells. We used DR-negative, DM-positive B-LCL expressing either DQ1, DP4 (3.1.3), or DQ2, DP4 (9.22.3), both derived from the same parental B-LCL in several steps of mutagenesis and selection (Ref. 8 and E. D. Mellins, unpublished data). We conducted staining and FACS analyses using CerCLIP, an Ab that recognizes CLIP peptides, bound to class II (14). DP4-CLIP complexes are not surface expressed, even in the absence of DM (see B-LCL 5.2.4; Fig. 1D). Thus, CerCLIP staining of B-LCLs 3.1.3 and 9.22.3 reflects DQ-CLIP complexes. DQ2-CLIP complexes were present at substantially higher levels than DQ1-CLIP complexes (with normalization for differences in levels of DQ) at the surface of DM-expressing cells (Fig. 1D). This result also implies DQ2-CLIP resistance to DM action.

DQ2 binds two CLIP sequences in different registers

To identify the specific CLIP peptides associated with DQ2, we eluted peptides from DQ2 molecules purified from the DM-expressing B-LCL 8.1.6 and analyzed them by MALDI-TOF and LC-coupled electrospray ionization mass spectrometry (MS). We confirmed the previous observation that Ii-derived peptides are associated with DQ2 expressed in B-LCL, even in the presence of DM (data not shown) (3, 4). Traces of CLIP are found among the peptides eluted from many, although not all, class II alleles expressed in B-LCL (15). The proportion of CLIP peptides in the total eluate from DQ2 is notably high, however. We also confirmed the prior finding that two families of CLIP peptides are associated with DQ2 from B-LCL: conventional CLIP peptides, here termed CLIP1, and unusual CLIP peptides, CLIP2 (see Tables II and III).


View this table:
[in this window]
[in a new window]

  		Table II. Identification of the Ii peptides bound to DQ2 and DR3a

 

View this table:
[in this window]
[in a new window]

  		Table III. Quantification of the Ii peptides bound to DQ2 and DR3a

 
CLIP1 peptides are derived from a sequence around Ii 83–101 and contain the Ii 91–99 core-binding motif. CLIP2 sequences, from around Ii 92–107, overlap with, but are C-terminal to, the sequences of CLIP1 peptides. The sequence differences in CLIP1 vs CLIP2 suggested that the peptides bound DQ2 in different frames. CLIP1 has been found to bind in the same frame to all alleles where this has been examined, with M91 in the P1 pocket (16). To identify the binding register of CLIP2 peptides to DQ2, we examined the binding of five 11-mers (Ii 97–107, Ii 96–106, Ii 95–105, Ii 94–104, Ii 93–103) that scan through Ii in positions 93–107 in comparison to a naturally processed form of CLIP2 peptide (Ii 92–107: RMATPLLMQALPMGAL) (4). We used a peptide-binding assay that measures the IC50 value of the test peptide, which is the amount of peptide needed to give 50% inhibition of the binding of a high-affinity indicator peptide. This analysis demonstrated that Ii 94–104 and Ii 92–107 had similarly low IC50 values, whereas Ii 95–105 and Ii 93–103 had increased IC50 values, consistent with weaker binding (Fig. 2A). This pattern implied that the core-binding region of CLIP2 was Ii 96–104 (PLLMQALPM), with P96 at the P1 position. This conclusion is compatible with the observation that N-terminally extended peptides bind with increased affinity due to hydrogen bond formation between the MHC II main chain and P-1 and P-2 (17).


Figure 2
View larger version (21K):
[in this window]
[in a new window]

  		FIGURE 2. The binding frame of CLIP2 for binding to DQ2 is determined to be PLLMQALPM. A, The IC50 values of truncated and mutated versions of the CLIP2 peptide were measured. IC50 is the peptide concentration required to give 50% inhibition of the indicator peptide KPLLIIAEDVEGEY. The underlined residues P and M denote positions P1 and P9 in the DQ2 peptide-binding frame. Dotted bars indicate IC50 values higher than 167 µM. Data are from one of at least three independent experiments. B, Depiction of the CLIP1- and CLIP2-binding frames.

 
To confirm the binding frame, we conducted a competitive inhibition binding assay with peptide variants of Ii 94–104, in which a lysine (K) was introduced in the putative P4, P5, or P6 positions (Ii 94–104;M99K, Ii 94–104;Q100K, and Ii 94–104;A101K, respectively). A positively charged lysine would be expected to prohibit binding to DQ2 when introduced at P4 and P6, but not at P5, as the P5 side chain is predicted to face the solvent. As seen in Fig. 2A, the IC50 value of Ii 94–104;Q100K is low, while the Ii 94–104;M99K and Ii 94–104;A101K values are high, consistent with our conclusion that PLLMQALPM (Ii 96–104) represents the core-binding region of CLIP2. The overlapping, but distinct, binding frames of CLIP1 and CLIP2 are shown schematically in Fig. 2B.

High levels of DQ2-Ii complexes in the presence or absence of DM

The finding of abundant CLIP peptides associated with DQ2 in cell lysates of DM-expressing cells led us to ask whether the DQ2-CLIP complexes found in DM-expressing cells represent a novel repertoire. We compared the peptide elution profile of DQ2 purified from 8.1.6 cells (DM expressing) with the elution profile of DQ2 purified from the DM-deficient mutant 9.5.3 derived from 8.1.6. MS analysis revealed that, like DQ2 from 8.1.6 cells, DQ2 from DM-deficient 9.5.3 cells also bound predominantly Ii-derived peptides, representing both CLIP1 and CLIP2 species. In contrast, and as previously reported (18), peptides eluted from DR3 (and DR52a) were predominantly Ii peptides in the absence of DM, whereas a heterogeneous mixture of self-peptides was associated with DR3 (and DR52a) from DM-expressing cells (data not shown). Since DR52a has very low affinity for CLIP (19), the observed CLIP peptides most probably derive from DR3. An overview of the Ii variants associated with DQ2 and DR3 is presented in Table II.

To more precisely quantify the CLIP peptide pools associated with DR3 and DQ2 in the presence and absence of DM, both the eluted peptides and synthetic CLIP peptides of a known amount were ICPL-labeled and analyzed by nano-LC-coupled Q-TOF-MS. The peak heights for the light-labeled eluted peptides were compared with those obtained for the heavy-labeled synthetic peptides. The results demonstrated that DR3 from DM-deficient cells had a 25-fold higher level of CLIP peptides compared with the DM-expressing parental line 8.1.6, while DQ2 from the DM-deficient cells showed only a 1.2-fold increase in total CLIP level compared with DQ2 from 8.1.6 (Fig. 3). This finding corroborated that DQ2-CLIP was significantly less affected by DM than DR3-CLIP.


Figure 3
View larger version (26K):
[in this window]
[in a new window]

  		FIGURE 3. Peptide exchange of DQ2 is relatively DM insensitive in vivo and in vitro. Quantification of eluted CLIP peptides from 50 µg of affinity-purified DQ2 and DR3 from 8.1.6 and 9.5.3 B-LCLs using LC-coupled Q-TOF-MS and ICPL of eluted and indicator peptides. The mean and range of at least two sets of independent experiments are shown. The quantities of the different Ii chain peptides and the electrospray ionization-MS/MS data justifying the selected peptides can be found in Table III.

 
Spontaneous peptide exchange is equivalent on DQ2 from DM+ or DM– cells

We next investigated the effect of the naturally processed peptide repertoires associated with DR3 and DQ2 on peptide exchange. We performed solid-phase peptide-binding assays in which DQ2 and DR3 molecules purified from lysates of 8.1.6 (DM+) and 9.5.3 (DM–) cells were captured by mAbs 2.12.E11 (anti-DQ2) or L243 (anti-DR dimer), and binding of biotinylated peptides to DQ2 or DR3 was measured. The spontaneous peptide exchange on DR3 originating from DM-null cells was increased >2-fold compared with DR3 from DM-expressing cells, reflecting the effects of DM editing of the peptide repertoire for less-exchangeable peptides. In contrast, peptide exchange on DQ2 from DM-deficient cells was similar to that observed with DQ2 from the DM-sufficient counterpart (Fig. 4A). These results are consistent with the similarity in the peptide profiles of DQ2 from 8.1.6 and 9.5.3 cells.


Figure 4
View larger version (16K):
[in this window]
[in a new window]

  		FIGURE 4. DQ2 displays reduced DM-mediated peptide exchange compared with DR3. A, Lysates of 8.1.6 and 9.5.3 B-LCLs were added to anti-DQ2 (2.12.E11)- or anti-DR (L243)-coated plates, followed by 48-h incubation with high-affinity biotinylated indicator peptides (EPRAPWIEQEGPEYW for DQ2 and KTIAYDEEARR for DR3). The peptide exchange of DQ2 or DR3 from 8.1.6 (wild type) is assigned a value of 1. Mean and SD from four independent experiments are shown. B, Peptide exchange of cleaved sDQ2-CLIP1, sDQ2-CLIP2, and sDR3-CLIP1 (1.5 µM) with high-affinity indicator peptides (1–5 nM KPLLIIAEDVEGEY for DQ2 and KTIAYDEEARR for DR3) were measured in the absence (0 DM) or presence of an increasing amount of DM (0.001–7.2 µM) using the fluid-phase peptide-binding assay. The complexes of DQ2 or DR3 with indicator peptides were used as a measure of peptide exchange. The negative control (without DQ2/DR3) did not give any signal (data not shown). The mean and range of two independent experiments are shown.

 
DQ2-CLIP1 and DQ2-CLIP2 display high resistance to DM action

To more precisely quantify the extent of DM resistance of DQ2-CLIP compared with DR3-CLIP for peptide exchange, we measured DM-mediated peptide exchange using a fluid-phase peptide exchange assay. Recombinant complexes (sDR3-CLIP1, sDQ2-CLIP1, sDQ2-CLIP2) were initially generated with tethered peptides attached to the N terminus of the class II β-chain with a linker including a thrombin cleavage site (see Materials and Methods). Thrombin-treated, sHLA molecules were incubated with high-affinity, allele-specific indicator peptides and increasing DM concentrations before measurement of binding of the indicator peptides. This in vitro assay allowed us to increase the DM:DQ2 ratio beyond that achieved in cells (including the DM transfectants). sDR3-CLIP1 was significantly more sensitive to DM, showing 50% of maximum peptide exchange at 10- and 20-fold lower concentrations of DM compared with sDQ2-CLIP1 and sDQ2-CLIP2, respectively (Fig. 4B).

DQ2 shows reduced binding interaction with DM compared with DQ1

To further explore the interaction of DQ2 and DM, we compared DQ2 and DQ1 molecules using an in vitro association assay in which bead-bound, immunoprecipitated DQ was incubated with soluble recombinant DM. DQ/DM binding is determined by isolating bead-associated DQ-DM complexes and detecting DM by immunoblotting of boiled complexes. With similar input amounts of DQ1 and DQ2 by silver stain (data not shown) and anti-class II β-chain blotting (Fig. 5A, upper panels), reduced amounts of soluble recombinant DM were coprecipitated with DQ2 (Fig. 5A, lower panels). The observed binding is specific, as DM binding to the Ab-conjugated beads alone is negligible (Fig. 5A, lower panels). DM/class II coprecipitation assays are known to be influenced by the peptide cargo of class II. Peptide-free class II appears to be the best ligand for DM, so that binding in this assay does not provide information that is independent of the ability of DM to release peptide from the class II-peptide complex (20). Accordingly, DQ1 from DM-null cells binds better than DQ1 from DM-positive cells, a portion of which is peptide edited (Fig. 5B). Notably, however, DM-DQ2 interaction is even more inefficient (Fig. 5B). We did not compare DR3 and DQ2 here, because of the confounding variable introduced by immunoprecipitating class II with different Abs.


Figure 5
View larger version (45K):
[in this window]
[in a new window]

  		FIGURE 5. DQ2 shows reduced binding interaction with DM compared with DQ1. DQ2 and DQ1 were immunoprecipitated with SPV-L3 from DM-positive (DR-null) B-LCLs 9.22.3 (DQ2) or 3.1.3 (DQ1) and DM-null cells 9.5.3 (DQ2) and 2.2.93 (DQ1), respectively, and then incubated with soluble recombinant DM to allow molecular interaction. A, Input class II amount was measured by immunoblotting with anti-class II β-chain Ab XD5.A11 (top) followed by measuring the amount of interacting DM by immunoblotting with anti-DM{alpha} mAb 5C1. Shown is one representative image of at least three experiments. B, Ratio of sDM:class II band intensity was calculated by densitometric analysis of the images in A: DM+ cell lines: 3.1.3 (DQ1) and 9.22.3 (DQ2) and DM-null cells: 2.2.93 (DQ1) and 9.5.3 (DQ2).

 
DQ2-CLIP1/2 complexes show high intrinsic stability and reduced DM effect

Effects of DM on various class II-peptide complexes and changes in the peptide repertoires of cells expressing or lacking DM have led to the conclusion that intrinsically more stable complexes are, in general, more resistant to DM (21, 22). The differential effect of DM on DQ2-CLIP and DR3-CLIP could reflect differences in the intrinsic stability of these complexes. Alternatively, the differences in peptide exchange could reflect reduced efficiency of DM interaction with the DQ2 allele. These two possibilities are not mutually exclusive. To assess the intrinsic stability of the DR3 and DQ2 complexes with CLIP peptides, we measured the spontaneous dissociation rate of the CLIP1 and CLIP2 peptides from sDQ2 and CLIP1 from sDR3 at pH 5 without DM using thrombin-treated, soluble molecules. The dissociation was measured by eluting the bound peptides and quantifying them by MALDI-TOF MS analysis, comparing the intensity of the isotopic peaks to an added indicator peptide. The CLIP1 and CLIP2 peptides from sDQ2 in the absence of DM displayed nearly identical dissociation rates (t1/2 = ~140 h), compared with the 2-fold faster release of CLIP1 from DR3 (t1/2 = ~61.5 h). Thus, both CLIP complexes with sDQ2 are more stable than sDR3-CLIP1, raising the possibility that increased intrinsic stability makes a contribution to the reduced efficacy of DM-mediated peptide exchange of the DQ2-CLIP complexes compared with DR3-CLIP1. In the presence of DM, using a class II:DM molar ratio of 1:3.8, the dissociation of CLIP1 from sDR3 increased 246 times (t1/2 = 0.25 h). DM also accelerated the release of CLIP1 and CLIP2 from sDQ2 (t1/2 = 9.0 h and t1/2 = 5.0 h, respectively); however, the DM-mediated enhancements of CLIP1/2 dissociations from DQ2 were modest (16–28 times).

DM effect on other DQ2-peptide complexes is also reduced

To explore the possibility that altered DM-DQ2 interaction contributed to the modest effect of DM on DQ2-CLIP1/2 half-lives, we assessed four additional sDQ2-peptide complexes; sDQ2-{gamma}I gliadin, sDQ2-{alpha}I gliadin, and two mutated forms of CLIP1 where the residue buried in MHC class II pocket 1 is mutated from methionine to either alanine (M91A) or proline (M91P). Compared with sDR3-CLIP1, the CLIP1 M91P peptide had a slightly longer spontaneous dissociation time (t1/2 = 93.6 h), the {gamma}I gliadin peptide had a similar dissociation rate (t1/2 = 55.2 h), while the intrinsic stability of the CLIP1 M91A and {alpha}I gliadin peptides were both lower than sDR3-CLIP1 (t1/2 = 13.1 h and 8.3 h, respectively). In the presence of DM, all four of these sDQ2-peptide complexes showed only moderate enhancement in peptide dissociation (5–19 times).

To further evaluate the DM susceptibility of these seven complexes, we sought to assess DM effects in relationship to their intrinsic stability. We compared our current results to previous findings where we measured peptide dissociation in the presence and absence of DM for a panel of complexes of varying intrinsic stability. In this prior work, we observed a consistent quantitative relationship between intrinsic stability and DM effect with a correlation of r = 0.69 (21). The sDR3-CLIP1 complex in the current study showed the predicted behavior, whereas, strikingly, all six of six sDQ2-peptide complexes did not conform to this quantitative relationship (Fig. 6). There are two other complexes with reductions in DM effect comparable to the sDQ2 complexes: one exceptionally stable complex (sDR0401-gp39 262–276) and a rapidly dissociating complex (DR1501-CLIP L97A variant). Our data predict that, under our experimental conditions, we also would see no/very low DM effect on very low stability complexes (t1/2 < 1.24 min) However, all tested complexes in the intrinsic stability range of the DQ2 complexes showed susceptibility to DM effect, and no other allele tested appears to have a consistent effect on DM susceptibility. These findings argue that the stability of the sDQ2-CLIP1/2 complexes in the presence of DM is not solely the result of their intrinsic stability.


Figure 6
View larger version (17K):
[in this window]
[in a new window]

  		FIGURE 6. Relationship between intrinsic and DM-catalyzed dissociation. Dissociation rate constants, k, calculated from the t1/2 by the equation t1/2 = ln2/k, are plotted on a log10 scale in units of hours–1. The x-axis shows dissociation in the absence of DM; the y-axis shows dissociation in the presence of sDM. The solid best-fit straight line through the data has a slope of 0.47 (95% confidence interval, 0.3–0.64; thin dashed curves): the extremely stable DR*0401 complex and the complexes studied in this report were excluded from the correlation analysis. The solid line of slope 1 through the origin indicates no DM effect; the vertical distance of each data point from this line is a measure of DM susceptibility. These data have been published with the exception of the data from the DQ2 and DR3 complexes (21 ). To compare current results to previous work on 36 complexes, it was necessary to correct values for the differences in the concentration of DM used; previous work indicated that effects of DM in the DM concentration range of both of these assays are linear (E. D. Mellins, unpublished data).

 
Cell surface expression of DQ2 is not enhanced by coexpression of DM

In addition to its peptide exchange function, DM acts as a chaperone and influences the steady state and cell surface abundance of certain alleles (Refs. 23, 24, 25, 26, 27 and S. Roh, C. H. Rinderknecht, A. Pashine, N. Lu, N. S. Patil, M. P. Belmares, T. Yoon, R. Busch, E. D. Mellins, submitted for publication). To see whether DQ2 is susceptible to these other effects of DM, we measured cell surface binding of a DQ2-specific Ab (2.12.E11) and of two monomorphic anti-DQ Abs to the DM-null cell 9.5.3 and its DM-transfectant, 9.5.3-DM. At saturating amounts of Ab, the binding of each of these Abs was comparable in the DM-expressing and nonexpressing cells, suggesting that cell surface abundance of DQ2 was not influenced by DM. In contrast, the level of DQ1 expression in the DM-null B-LCL 2.2.93 was increased by expression of DM in the 2.2.93 transfectant. The difference in effects on DQ1 and DQ2 was not due to cell line differences, because the behavior of DP4 and DR3 molecules was consistent across DM transfectants of both cell lines (Fig. 7).


Figure 7
View larger version (21K):
[in this window]
[in a new window]

  		FIGURE 7. Cell surface expression level of DQ2 is not enhanced by coexpression of DM. Fold change in MFI of staining of 2.2.93-DM compared with 2.2.93 (anti-DR (L243), anti-DP (B7/21.2), anti-DQ (SPV-L3 and aIa3) staining), or of 9.5.3-DM compared with 9.5.3 (anti-DR (L243), anti-DP (B7/21.2), anti-DQ (SPV-L3 and aIa3), anti-DQ2 (b2.12.E11), and anti-DR3 (c16.23) staining). Mean values from multiple experiments are shown as horizontal bars. Dotted line indicates no change (MFI ratio = 1). Analyses using t tests showed that all of the changes except DQ2a, DQ2b, and DR3 are significant (p < 0.05).

 
Conformation of DQ1 but not DQ2 is altered by DM expression

DM interaction with nascent class II molecules influences their conformation, probably in both peptide-dependent and -independent manners (28). We assessed whether DQ2 was susceptible to DM effects on molecular maturation, in comparison to DQ1. After pulse-chase metabolic labeling of the same cell lines used in Fig. 7, we immunoprecipitated DQ1 and DQ2 with the monomorphic anti-DQ Ab SPV-L3. We observed increased binding of SPV-L3 to DQ1 at ~2.5 h of chase in 2.2.93-DM, but not 2.2.93 cells, consistent with a DM-dependent conformational change in DQ1 (Fig. 8). In contrast, DQ2 (expressed in 9.5.3 with or without DM) did not show the DM-dependent conformational change detected by SPV-L3 (Fig. 8) or the DQ2-specific Ab 2.12.E11 (data not shown). This result does not depend on the cell lines tested since DP4 undergoes DM-dependent increases in precipitable molecules in 9.5.3-DM and 2.2.93-DM (data not shown; Roh et al., submitted for publication). We conclude that DQ2 resists DM-mediated chaperoning as well as peptide exchange.


Figure 8
View larger version (41K):
[in this window]
[in a new window]

  		FIGURE 8. DQ2 does not show DM-dependent conformational change during biosynthesis. 2.2.93, 9.5.3, and their DM transfectants (2.2.93-DM and 9.5.3-DM, respectively) were pulsed for 15 min with [35S]methionine/cysteine and then chased for 2, 2.5, 3, 3.5, 4, and 4.5 h (lanes 1, 2, 3, 4, 5, 6, respectively) as described in Materials and Methods. Cell lysates (5 x 106 cell equivalents/lane) were immunoprecipitated with anti-DQ mAb SPV-L3 and then analyzed by SDS-PAGE. Representative images from one of five independent experiments are shown. *, (1 = 2.2.93; 2 = 2.2.93-DM; 3 = 9.5.3; and 4 = 9.5.3-DM) at 0 h.

 
The interaction of DM with DQ2 is competitively inhibited by DR3

Given that DR3 and DQ2 are tightly linked alleles and are generally expressed concurrently, they likely are competing ligands for DM within peptide-loading compartments. To model this situation in vitro, we examined whether DR3 could competitively inhibit DQ2-DM interaction. We used sDR3 or sDQ2 with tethered CLIP1 as a competitor for DQ2 or DR3, respectively, with thrombin-cleaved CLIP1, in a DM-mediated peptide exchange reaction. We varied the ratios of thrombin-treated to thrombin-untreated complexes and measured binding of labeled, high-affinity peptides. At equimolar, 2- and 10-fold molar excess concentrations of DR3, the peptide exchange on DQ2 fell by 15, 27. and 61%, respectively. DQ2 was a less effective inhibitor, reducing the peptide exchange on DR3 by 2, 6, and 38% at the same concentrations (Fig. 9). These results argue that DR3 is a more potent inhibitor of DM-mediated peptide exchange of DQ2 than vice versa. At 10-fold molar excess of DR3 compared with DQ2, which roughly approximates physiological intracellular conditions (29), the inefficient DQ2-CLIP-DM interaction is further hampered by the presence of a successful, competitive ligand for DM.


Figure 9
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 9. DQ2-DM interaction is subject to competitive inhibition by DR3. The level of peptide exchange of cleaved sDQ2-CLIP1 or sDR3-CLIP1 (0.7 µM) with indicator peptides (EPRAPWIEQEGPEYW for DQ2 and KTIAYDEEARR for DR3) was measured in the presence of DM (0.3 µM) and varying concentrations of noncleaved sDR3-CLIP1 or sDQ2-CLIP1 (0–7 µM). Mean and SD of three independent experiments are shown.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
For most MHC class II alleles, biosynthesis and peptide loading is regulated by the actions of two cofactors: Ii and DM. Full-length Ii typically associates with class II with its core CLIP region (CLIP1; Ii 91–99) bound to the class II peptide-binding groove, and nested peptides from this region are left in the groove after Ii proteolysis. DM typically binds to class II to catalyze CLIP release, regulate class II conformation, and edit the repertoire of bound peptides that will be presented to CD4+ T cells (30, 31, 32). The class II molecule DQ2 is associated with unusually high amounts of two CLIP variants (CLIP1 and CLIP2), and we find that the CLIP-rich phenotype of DQ2 in cells appears nearly identical in the presence and absence of DM. This unusual phenotype of DQ2 is apparently explained by a combination of the relatively high intrinsic stability of the DQ2-CLIP1/2 complexes and the inefficient interaction of DQ2 with DM.

Formally, DM resistance of DQ2-CLIP could result from a thermodynamic block (reduced DM binding to DQ2-CLIP) or a kinetic block (reduced ability of DM to release CLIP from DQ2). We found that increased concentrations of DM result in more effective CLIP release from DQ2, as shown by the reduction of coprecipitated CLIP in the presence of DM and by peptide exchange assays, arguing that the block is thermodynamic. However, much higher concentrations of DM are needed to obtain a peptide-editing effect equivalent to that observed with DR3-CLIP1. The higher DM concentrations or prolonged exposure required for activity with DQ2 substrates may rarely be achieved in vivo for several reasons. DM levels are substoichiometric to class II in peptide-loading compartments, with the DM:DR molar ratio {approx}1:5 (Ref. 33 and E. D. Mellins, unpublished data). Endosomal residence time for class II-CLIP complexes is seemingly kinetically controlled, because DR3-CLIP complexes arrive at the cell surface at a rate that is similar to DR3-peptide complexes; prolonged retention of either class II-CLIP or more DM-resistant complexes does not appear to occur (34). With linkage disequilibrium in MHC class II haplotypes and coregulated expression of class II isotypes, DQ2 is almost always expressed concurrently with DR3, which is capable of out-competing DQ2 for limiting amounts of DM, as evident in Fig. 9.

The structural basis of the unusual interaction of DQ2 with DM is likely to involve polymorphic residues of DQ2. DM is thought to disrupt H bonds between the peptide backbone and residues around pocket 1 of the class II molecule ({alpha}53, β81) or stabilize a class II conformation in which these hydrogen bonds are disrupted, along with more global changes effecting pocket structure (7, 35, 36, 37, 38, 39, 40). Mutational analysis has demonstrated the critical role of the protruding {alpha}51F in DR3 for interaction with DM (37, 38). This residue is predicted to participate in hydrophobic interactions with DM and may serve as a lever to alter the location of the extended strand including {alpha}51–53 (37, 38) and its distance from the bound peptide. In DQ2, the {alpha}44–53 stretch (numbering as in Ref. 41) contains several polymorphic residues and deletion of {alpha}53. Further work should identify whether this region indeed is of significance and which residues are involved.

It is striking that CLIP2 is a prominent component of the DQ2 peptide repertoire. We show that CLIP2 binds to DQ2 by placing a proline at the P1 position. As part of a polypeptide chain, the proline side chain is unable to engage in amide hydrogen bonding because of its ring structure. In most MHC class II molecules, there is a hydrogen bond from residue 53 of the {alpha}-chain to the amide nitrogen of the P1 residue. Not only does DQ2 have a deletion of residue {alpha}53, but there is evidence that DQ2 is unable to form the P1-related hydrogen bond (42). Thus, proline can be accommodated by DQ2 at P1 without energetic penalty, perhaps explaining the unusual association of DQ2 with CLIP2 peptides. An interesting question is how these CLIP2 fragments come to be associated with DQ2. The CLIP region is unstructured within the full-length Ii and may allow an alternate register of association in the endoplasmic reticulum. Alternatively, CLIP2 peptide associates with DQ2 after some proteolysis of Ii in endosomes, either through reptation or rebinding.

On first glance, it may appear that our data on the DM susceptibility of DQ2-CLIP1 compared with DQ2-CLIP2 complexes are inconsistent. Among naturally processed peptides, the proportion of DQ2-CLIP2 is modestly increased in DM-expressing compared with nonexpressing cells, suggesting increased resistance to DM editing. In the peptide exchange assay (Fig. 4B), the dose-response curve with DM is shifted to the right for sDQ2-CLIP2 compared with sDQ2-CLIP1. However, in Fig. 6, off-rate measurements of the two forms of CLIP demonstrate comparable values in the absence of DM and a longer half-life of the sDQ2-CLIP1 complex in the presence of DM. These findings can be rationalized, because the data in Fig. 4B reflect the net effect of off-rate and on-rate of the CLIP variants in the presence of DM, while the assay in Fig. 6 includes excess competitor peptide, precluding any rebinding of the CLIP peptides and therefore any influence of peptide on-rates.

HLA haplotypes with the DR3-DQ2 segment are associated with many autoimmune disorders, including celiac disease, type 1 diabetes, systemic lupus erythematosus, Graves’ disease, and Addison’s disease (5, 6). Notably, for many of these diseases, the major HLA effects have been mapped to the DR3-DQ2 interval (43, 44). The direct comparison of DR3 and DQ2, as we have done, is therefore of particular interest. The unusual interactions between DQ2 and Ii and DM could be one reason for the widespread predisposition to autoimmunity of DR3-DQ2 haplotypes. The DQ2-related effects may act at two levels: at the level of presentation of peptide epitopes in the periphery and at the level of T cell selection in the thymus.

In the periphery, immunodominant T cell epitopes are mainly selected based on the (high) kinetic stability of the peptide-MHC II interaction (45, 46, 47), a process generally enhanced by the presence of HLA-DM (47, 48, 49). Exceptions have been found in autoimmunity models, where peptides with low kinetic stability proved immunodominant (46, 50, 51, 52). The presentation of these low-affinity peptides may be augmented by a reduced DM interaction and by being in high abundance. The vigorous immune responses of celiac disease patients to gluten peptides that bind DQ2 with relatively low affinity can thus relate to both reduced DM interaction of DQ2 and the presence of high amounts of gluten peptides in the gut mucosa.

In the thymus, the overabundance of unexchanged CLIP peptides may affect T cell selection. Studies in mice have demonstrated that a narrow spectrum of MHC class II ligands profoundly alters the repertoire of CD4+ T cells (53, 54, 55, 56). The effects include reduction in both number and diversity of CD4+ T cells, thought to be due to inefficiency of thymic-positive selection (53, 54). In addition, T cells within this population are negatively selected on a limited self-peptide repertoire and show broad reactivity to self-peptides, increasing their potential for autoreactivity (54, 56).

Another aspect of escaping DM editing is that the residual CLIP1/2 peptides, although of relatively higher affinity for DQ2 than CLIP1 peptides for many other alleles, nonetheless may be less ideal than a DM-edited peptide repertoire for stabilizing class II surface expression and inhibiting exchange with extracellular peptides, including potential autoantigens (9, 57). Our previous work on the half-life of complexes that survive DM editing shows a 9- to 70-fold increase in intrinsic half-life compared with DR3-CLIP, whereas DQ2-CLIP has only a 2.3-fold increase in t1/2 compared with DR3-CLIP (58). This difference suggests DQ2-CLIP1/2 would remain exchangeable on the surface of DM+ cells, especially under conditions (e.g., peptide availability and pH) that favor exchange. In addition to these characteristics of DQ2, which may create a general predisposition to a breakdown in self-tolerance, unique features of DQ2 peptide binding may play a role in particular diseases. For example, we have hypothesized that DQ2 allows binding of proline-rich gluten epitopes of celiac disease, because it tolerates registers with proline residues at P1 (41, 42).

Why has DQ2 of the DR3-DQ2 haplotype evolved to be less DM sensitive? One possibility is that, under certain circumstances, this characteristic will lead to presentation of numerous antigenic epitopes instead of a few immunodominant epitopes (49), and this may provide a selective advantage in combating some infectious diseases. However, a deleterious side effect could be an increased susceptibility to autoimmune diseases. Studies in humanized mice models should shed light on the in vivo relevance of the phenomena described here.


   Acknowledgments
 
We thank Peter Cresswell for the CerCLIP and XD5.A11 Abs, Hergen Spits for the SPV-L3 Ab, and Bernard Malissen for the B8.11 Ab. We thank Khoa Nguyen (Mellins’ group) for assistance with figures.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by grants from the Research Council of Norway, the National Institutes of Health (U19 AI02042 to E.D.M.), the Wasie Foundation (to E.D.M.). and the Norwegian Foundation for Health and Rehabilitation. Back

2 E.D.M. and L.M.S. made equivalent contributions. Back

3 Address correspondence and reprint requests to Dr. Elizabeth D. Mellins, Department of Pediatrics, Stanford University, Stanford, CA 94305 and Dr. Ludvig M. Sollid, Institute of Immunology, Rikshospitalet, 0027 Oslo, Norway. E-mail addresses: mellins{at}stanford.edu and l.m.sollid{at}medisin.uio.no Back

4 Abbreviations used in this paper: Ii, invariant chain; ACN, acetonitrile; B-LCL, B lymphoblastoid cell line; s, soluble; TFA, trifluoroacetic acid; MFI, mean fluorescence intensity; MS, mass spectrometry; ICPL, isotope coded peptide labeling; LC, liquid chromatography. Back

Received for publication August 7, 2007. Accepted for publication August 16, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Cresswell, P.. 1994. Assembly, transport, and function of MHC class II molecules. Annu. Rev. Immunol. 12: 259-293. [Medline]
  2. Busch, R., C. H. Rinderknecht, S. Roh, A. W. Lee, J. J. Harding, T. Burster, T. M. Hornell, E. D. Mellins. 2005. Achieving stability through editing and chaperoning: regulation of MHC class II peptide binding and expression. Immunol. Rev. 207: 242-260. [Medline]
  3. van de Wal, Y., Y. M. Kooy, J. W. Drijfhout, R. Amons, F. Koning. 1996. Peptide binding characteristics of the coeliac disease-associated DQ({alpha}1*0501, β1*0201) molecule. Immunogenetics 44: 246-253. [Medline]
  4. Vartdal, F., B. H. Johansen, T. Friede, C. J. Thorpe, S. Stevanovic, J. E. Eriksen, K. Sletten, E. Thorsby, H. G. Rammensee, L. M. Sollid. 1996. The peptide binding motif of the disease associated HLA-DQ ({alpha}1*0501, β1*0201) molecule. Eur. J. Immunol. 26: 2764-2772. [Medline]
  5. Lechler, R., A. Warrens. 2000. HLA in Health and Disease Academic Press, London.
  6. Price, P., C. Witt, R. Allcock, D. Sayer, M. Garlepp, C. C. Kok, M. French, S. Mallal, F. Christiansen. 1999. The genetic basis for the association of the 8.1 ancestral haplotype (A1, B8, DR3) with multiple immunopathological diseases. Immunol. Rev. 167: 257-274. [Medline]
  7. Pashine, A., R. Busch, M. P. Belmares, J. N. Munning, R. C. Doebele, M. Buckingham, G. P. Nolan, E. D. Mellins. 2003. Interaction of HLA-DR with an acidic face of HLA-DM disrupts sequence-dependent interactions with peptides. Immunity 19: 183-192. [Medline]
  8. Pious, D., L. Dixon, F. Levine, T. Cotner, R. Johnson. 1985. HLA class II regulation and structure: analysis with HLA-DR3 and HLA-DP point mutants. J. Exp. Med. 162: 1193-1207. [Abstract/Free Full Text]
  9. Mellins, E., S. Kempin, L. Smith, T. Monji, D. Pious. 1991. A gene required for class II-restricted antigen presentation maps to the major histocompatibility complex. J. Exp. Med. 174: 1607-1615. [Abstract/Free Full Text]
  10. Quarsten, H., S. N. McAdam, T. Jensen, H. Arentz-Hansen, O. Molberg, K. E. Lundin, L. M. Sollid. 2001. Staining of celiac disease-relevant T cells by peptide-DQ2 multimers. J. Immunol. 167: 4861-4868. [Abstract/Free Full Text]
  11. Sloan, V. S., P. Cameron, G. Porter, M. Gammon, M. Amaya, E. Mellins, D. M. Zaller. 1995. Mediation by HLA-DM of dissociation of peptides from HLA-DR. Nature 375: 802-806. [Medline]
  12. Johansen, B. H., S. Buus, F. Vartdal, H. Viken, J. A. Eriksen, E. Thorsby, L. M. Sollid. 1994. Binding of peptides to HLA-DQ molecules: peptide binding properties of the disease-associated HLA-DQ ({alpha}1*0501, β1*0201) molecule. Int. Immunol. 6: 453-461. [Abstract/Free Full Text]
  13. Vader, W., D. Stepniak, Y. Kooy, L. Mearin, A. Thompson, J. J. van Rood, L. Spaenij, F. Koning. 2003. The HLA-DQ2 gene dose effect in celiac disease is directly related to the magnitude and breadth of gluten-specific T cell responses. Proc. Natl. Acad. Sci. USA 100: 12390-12395. [Abstract/Free Full Text]
  14. Avva, R. R., P. Cresswell. 1994. In vivo and in vitro formation and dissociation of HLA-DR complexes with invariant chain-derived peptides. Immunity 1: 763-774. [Medline]
  15. Chicz, R. M., R. G. Urban, J. C. Gorga, D. A. Vignali, W. S. Lane, J. L. Strominger. 1993. Specificity and promiscuity among naturally processed peptides bound to HLA-DR alleles. J. Exp. Med. 178: 27-47. [Abstract/Free Full Text]
  16. Ghosh, P., M. Amaya, E. Mellins, D. C. Wiley. 1995. The structure of an intermediate in class II MHC maturation: CLIP bound to HLA-DR3. Nature 378: 457-462. [Medline]
  17. Sant, A. J., C. Beeson, B. McFarland, J. Cao, S. Ceman, P. W. Bryant, S. Wu. 1999. Individual hydrogen bonds play a critical role in MHC class II: peptide interactions: implications for the dynamic aspects of class II trafficking and DM-mediated peptide exchange. Immunol. Rev. 172: 239-253. [Medline]
  18. Mellins, E., P. Cameron, M. Amaya, S. Goodman, D. Pious, L. Smith, B. Arp. 1994. A mutant human histocompatibility leukocyte antigen DR molecule associated with invariant chain peptides. J. Exp. Med. 179: 541-549. [Abstract/Free Full Text]
  19. Sette, A., S. Southwood, J. Miller, E. Appella. 1995. Binding of major histocompatibility complex class II to the invariant chain-derived peptide, CLIP, is regulated by allelic polymorphism in class II. J. Exp. Med. 181: 677-683. [Abstract/Free Full Text]
  20. Denzin, L. K., C. Hammond, P. Cresswell. 1996. HLA-DM interactions with intermediates in HLA-DR maturation and a role for HLA-DM in stabilizing empty HLA-DR molecules. J. Exp. Med. 184: 2153-2166. [Abstract/Free Full Text]
  21. Belmares, M. P., R. Busch, K. W. Wucherpfennig, H. M. McConnell, E. D. Mellins. 2002. Structural factors contributing to DM susceptibility of MHC class II/peptide complexes. J. Immunol. 169: 5109-5117. [Abstract/Free Full Text]
  22. Weber, D. A., B. D. Evavold, P. E. Jensen. 1996. Enhanced dissociation of HLA-DR-bound peptides in the presence of HLA-DM. Science 274: 618-621. [Abstract/Free Full Text]
  23. Albert, L. J., B. Ghumman, T. H. Watts. 1996. Effect of HLA-DM transfection on hen egg lysozyme presentation by T2.Ak cells. J. Immunol. 157: 2247-2255. [Abstract]
  24. Fallas, J. L., H. M. Tobin, O. Lou, D. Guo, D. B. Sant'Angelo, L. K. Denzin. 2004. Ectopic expression of HLA-DO in mouse dendritic cells diminishes MHC class II antigen presentation. J. Immunol. 173: 1549-1560. [Abstract/Free Full Text]
  25. Honey, K., K. Forbush, P. E. Jensen, A. Y. Rudensky. 2004. Effect of decreasing the affinity of the class II-associated invariant chain peptide on the MHC class II peptide repertoire in the presence or absence of H-2M. J. Immunol. 172: 4142-4150. [Abstract/Free Full Text]
  26. Koonce, C. H., G. Wutz, E. J. Robertson, A. B. Vogt, H. Kropshofer, E. K. Bikoff. 2003. DM loss in k haplotype mice reveals isotype-specific chaperone requirements. J. Immunol. 170: 3751-3761. [Abstract/Free Full Text]
  27. Lich, J. D., J. A. Jayne, D. Zhou, J. F. Elliott, J. S. Blum. 2003. Editing of an immunodominant epitope of glutamate decarboxylase by HLA-DM. J. Immunol. 171: 853-859. [Abstract/Free Full Text]
  28. Lovitch, S. B., Z. Pu, E. R. Unanue. 2006. Amino-terminal flanking residues determine the conformation of a peptide-class II MHC complex. J. Immunol. 176: 2958-2968. [Abstract/Free Full Text]
  29. Gorga, J. C., V. Horejsi, D. R. Johnson, R. Raghupathy, J. L. Strominger. 1987. Purification and characterization of class II histocompatibility antigens from a homozygous human B cell line. J. Biol. Chem. 262: 16087-16094. [Abstract/Free Full Text]
  30. Kropshofer, H., S. O. Arndt, G. Moldenhauer, G. J. Hammerling, A. B. Vogt. 1997. HLA-DM acts as a molecular chaperone and rescues empty HLA-DR molecules at lysosomal pH. Immunity 6: 293-302. [Medline]
  31. Kropshofer, H., A. B. Vogt, G. Moldenhauer, J. Hammer, J. S. Blum, G. J. Hammerling. 1996. Editing of the HLA-DR-peptide repertoire by HLA-DM. EMBO J. 15: 6144-6154. [Medline]
  32. van Ham, S. M., U. Gruneberg, G. Malcherek, I. Broker, A. Melms, J. Trowsdale. 1996. Human histocompatibility leukocyte antigen (HLA)-DM edits peptides presented by HLA-DR according to their ligand binding motifs. J. Exp. Med. 184: 2019-2024. [Abstract/Free Full Text]
  33. Schafer, P. H., J. M. Green, S. Malapati, L. Gu, S. K. Pierce. 1996. HLA-DM is present in one-fifth the amount of HLA-DR in the class II peptide-loading compartment where it associates with leupeptin-induced peptide (LIP)-HLA-DR complexes. J. Immunol. 157: 5487-5495. [Abstract]
  34. Guerra, C. B., R. Busch, R. C. Doebele, W. Liu, T. Sawada, W. W. Kwok, M. D. Chang, E. D. Mellins. 1998. Novel glycosylation of HLA-DR{alpha} disrupts antigen presentation without altering endosomal localization. J. Immunol. 160: 4289-4297. [Abstract/Free Full Text]
  35. Busch, R., A. Pashine, K. C. Garcia, E. D. Mellins. 2002. Stabilization of soluble, low-affinity HLA-DM/HLA-DR1 complexes by leucine zippers. J. Immunol. Methods 263: 111-121. [Medline]
  36. Chou, C. L., S. Sadegh-Nasseri. 2000. HLA-DM recognizes the flexible conformation of major histocompatibility complex class II. J. Exp. Med. 192: 1697-1706. [Abstract/Free Full Text]
  37. Doebele, R. C., R. Busch, H. M. Scott, A. Pashine, E. D. Mellins. 2000. Determination of the HLA-DM interaction site on HLA-DR molecules. Immunity 13: 517-527. [Medline]
  38. Mosyak, L., D. M. Zaller, D. C. Wiley. 1998. The structure of HLA-DM, the peptide exchange catalyst that loads antigen onto class II MHC molecules during antigen presentation. Immunity 9: 377-383. [Medline]
  39. Narayan, K., C. L. Chou, A. Kim, I. Z. Hartman, S. Dalai, S. Khoruzhenko, S. Sadegh-Nasseri. 2007. HLA-DM targets the hydrogen bond between the histidine at position β81 and peptide to dissociate HLA-DR-peptide complexes. Nat. Immunol. 8: 92-100. [Medline]
  40. Stratikos, E., D. C. Wiley, L. J. Stern. 2004. Enhanced catalytic action of HLA-DM on the exchange of peptides lacking backbone hydrogen bonds between their N-terminal region and the MHC class II {alpha}-chain. J. Immunol. 172: 1109-1117. [Abstract/Free Full Text]
  41. Kim, C. Y., H. Quarsten, E. Bergseng, C. Khosla, L. M. Sollid. 2004. Structural basis for HLA-DQ2-mediated presentation of gluten epitopes in celiac disease. Proc. Natl. Acad. Sci. USA 101: 4175-4179. [Abstract/Free Full Text]
  42. Bergseng, E., J. Xia, C. Y. Kim, C. Khosla, L. M. Sollid. 2005. Main chain hydrogen bond interactions in the binding of proline-rich gluten peptides to the celiac disease associated HLA-DQ2 molecule. J. Biol. Chem. 280: 21791-21796. [Abstract/Free Full Text]
  43. Nejentsev, S., J. M. M. Howson, N. M. Walker, J. Szeszko, S. F. Field, H. E. Stevens, P. Reynolds, M. Hardy, E. King, J. Masters, et al 2007. Localization of type 1 diabetes susceptibility to the MHC class I genes HLA-B and HLA-A. Nature 450: 887-892. [Medline]
  44. Fernando, M. M., C. R. Stevens, P. C. Sabeti, E. C. Walsh, A. J. McWhinnie, A. Shah, T. Green, J. D. Rioux, T. J. Vyse. 2007. Identification of two independent risk factors for lupus within the MHC in United Kingdom families. PLoS Genet. 3: 2109-2121.
  45. Lazarski, C. A., F. A. Chaves, S. A. Jenks, S. Wu, K. A. Richards, J. M. Weaver, A. J. Sant. 2005. The kinetic stability of MHC class II:peptide complexes is a key parameter that dictates immunodominance. Immunity 23: 29-40. [Medline]
  46. Hall, F. C., J. D. Rabinowitz, R. Busch, K. C. Visconti, M. Belmares, N. S. Patil, A. P. Cope, S. Patel, H. M. McConnell, E. D. Mellins, G. Sonderstrup. 2002. Relationship between kinetic stability and immunogenicity of HLA-DR4/peptide complexes. Eur. J. Immunol. 32: 662-670. [Medline]
  47. Lazarski, C. A., F. A. Chaves, A. J. Sant. 2006. The impact of DM on MHC class II-restricted antigen presentation can be altered by manipulation of MHC-peptide kinetic stability. J. Exp. Med. 203: 1319-1328. [Abstract/Free Full Text]
  48. Nanda, N. K., A. J. Sant. 2000. DM determines the cryptic and immunodominant fate of T cell epitopes. J. Exp. Med. 192: 781-788. [Abstract/Free Full Text]
  49. Nanda, N. K., E. K. Bikoff. 2005. DM peptide-editing function leads to immunodominance in CD4 T cell responses in vivo. J. Immunol. 175: 6473-6480. [Abstract/Free Full Text]
  50. Joosten, I., M. H. Wauben, M. C. Holewijn, K. Reske, L. O. Pedersen, C. F. Roosenboom, E. J. Hensen, W. van Eden, S. Buus. 1994. Direct binding of autoimmune disease related T cell epitopes to purified Lewis rat MHC class II molecules. Int. Immunol. 6: 751-759. [Abstract/Free Full Text]
  51. Seamons, A., J. Sutton, D. Bai, E. Baird, N. Bonn, B. F. Kafsack, J. Shabanowitz, D. F. Hunt, C. Beeson, J. Goverman. 2003. Competition between two MHC binding registers in a single peptide processed from myelin basic protein influences tolerance and susceptibility to autoimmunity. J. Exp. Med. 197: 1391-1397. [Abstract/Free Full Text]
  52. Fairchild, P. J., D. C. Wraith. 1996. Lowering the tone: mechanisms of immunodominance among epitopes with low affinity for MHC. Immunol. Today 17: 80-85. [Medline]
  53. Ignatowicz, L., J. Kappler, P. Marrack. 1996. The repertoire of T cells shaped by a single MHC/peptide ligand. Cell 84: 521-529. [Medline]
  54. Martin, W. D., G. G. Hicks, S. K. Mendiratta, H. I. Leva, H. E. Ruley, L. Van Kaer. 1996. H2-M mutant mice are defective in the peptide loading of class II molecules, antigen presentation, and T cell repertoire selection. Cell 84: 543-550. [Medline]
  55. Singh, N., L. Van Kaer. 1999. Unexpected reactivities of T cells selected by a single MHC-peptide ligand. J. Immunol. 163: 3583-3591. [Abstract/Free Full Text]
  56. Surh, C. D., D. S. Lee, W. P. Fung-Leung, L. Karlsson, J. Sprent. 1997. Thymic selection by a single MHC/peptide ligand produces a semidiverse repertoire of CD4+ T cells. Immunity 7: 209-219. [Medline]
  57. Mellins, E., L. Smith, B. Arp, T. Cotner, E. Celis, D. Pious. 1990. Defective processing and presentation of exogenous antigens in mutants with normal HLA class II genes. Nature 343: 71-74. [Medline]
  58. Doebele, R. C., A. Pashine, W. Liu, D. M. Zaller, M. Belmares, R. Busch, E. D. Mellins. 2003. Point mutations in or near the antigen-binding groove of HLA-DR3 implicate class II-associated invariant chain peptide affinity as a constraint on MHC class II polymorphism. J. Immunol. 170: 4683-4692. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
Z. Zhou, K. A. Callaway, D. A. Weber, and P. E. Jensen
Cutting Edge: HLA-DM Functions through a Mechanism That Does Not Require Specific Conserved Hydrogen Bonds in Class II MHC-Peptide Complexes
J. Immunol., October 1, 2009; 183(7): 4187 - 4191.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow A correction has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fallang, L.-E.
Right arrow Articles by Sollid, L. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fallang, L.-E.
Right arrow Articles by Sollid, L. M.
Right arrowPubmed/NCBI databases
Medline Plus Health Information
*Autoimmune Diseases

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS